The agroecological transition offer opportunities to reduce agriculture's environmental impacts by reducing reliance on synthetic fertilisers and pesticides. Crop diversification, in both time and space, is a key strategy including extended crop rotations, intercropping, and cover crops. Yet, relationships between reduced input use and associated environmental impacts remain insufficiently quantified. We assessed the environmental performance of six innovative low-input cropping systems that used cover crops, cultivar mixture and intercropping in term of nitrogen fertiliser and pesticide use, as well as nitrate and pesticide losses. From 2010 to 2016, cropping systems were monitored for input use and drainage water was collected with tension plate lysimeters at 1 m depth. Nitrate and up to 44 pesticide compounds were analysed annually. Nitrogen fertiliser application varied across systems, with more diversified systems applying less thanks to legumes. Pesticide use remained similar among systems but reduced by over 50% compared to conventional rates. Cover crops played a key role in reducing pollution. Nitrate leaching reduced by 42–56% in systems with cover crops. More originally, pesticide leaching decreased by 53–82% for these systems with S-metolachlor representing more than 50% of the quantity of pesticide losses. These results demonstrate that diversifying cropping systems, particularly through cover cropping, can reduce agriculture's environmental footprint greatly. When combined with reducing input use, such strategies provide a promising pathway towards more sustainable and resilient farming systems, with clear benefits for water quality and agroecosystem functioning. • Six low-input diversified arable cropping systems were tested for six years. • Nitrate and pesticide leaching were collected over the six years of monitoring. • Cover crops cut nitrate leaching by 42–56% under field conditions. • First evidence that cover crops reduce pesticide leaching by 53–82%. • S-metolachlor loss is the main driver of the pesticide leaching pattern observed.

The National Land Use Inventory (IUTI), implemented by the Ministry for Environment, Land and Sea Protection (now Ministry of Environment and Energy Security), is the main information source of the National Registry for forest Carbon Sinks. Over time, it has proven to be a valuable tool for investigating land use dynamics, even in its 1% subsample, statistically valid for many applications. This paper presents the evolution of the IUTI sampling system and analyzes land-use changes in Italy over the past 30 years, from 1990 to the 2022 update, carried out as part of the GeoSciencesIR Project. The emerging trends are consistent with those highlighted in the 1990-2008 land-use analysis and with the dynamics observed in European and Mediterranean contexts. The main trends highlighted: (i) a significant reduction in Arable land and other herbaceous crops, from 11.4 to 9.6 million ha, with a smaller contraction than in the first period. This land-use class was dominant in Italy until a few years ago; as of 2022, it has been surpassed in extent by woodland. The main cause of reduction is linked to the expansion of urbanized areas in the plains and the progressive recolonization of abandoned farmland by forest in hilly and mountainous areas. (ii) A significant reduction in Grasslands, pastures, and uncultivated land, from 2.1 to 1.7 million ha, mainly due to natural recolonization by woodland. This dynamic could lead to the progressive reduction in semi-natural areas in the future, with all the ensuing environmental and landscape implications. (iii) A considerable expansion in woodland, particularly in hilly and mountainous areas, from 9.1 to 9.9 million ha, becoming the dominant land use class. This expansion, with a lower rate in the second period (2008-2022), is mainly due to the natural evolution of “other wooded lands” and the recolonization of abandoned farmland and disused pastures, largely driven by the ongoing depopulation of mountainous areas. (iv) The progressive expansion of urbanized areas, from 1.6 to 2.3 million ha, especially in the lowland belt, mainly driven by urban sprawl, which still shows steady growth in Italy. Despite the many benefits brought by forest expansion, this spontaneous recolonization is contributing to the loss of cultural landscapes linked to agro-pastoral practices, as well as to the progressive reduction of open spaces and to the homogenization of the land. All these issues raise important questions in the definition of land-use planning strategies. In this regard, IUTI has proven to be a reliable information source for the analysis of land use dynamics, providing a solid reference for surveys at different scales.

Agriculture is currently facing rising global food demand, decreasing arable land availability, and intensifying climate-related stresses. Traditional farming practices rely heavily on chemical fertilizers and pesticides, leading to soil degradation, water contamination, and increased greenhouse gas emissions. To ensure sustainable food production, adopting eco-friendly, resource-efficient green technologies has become crucial. This study examines the combined impact of precision agriculture, advanced water management systems, and bio-based inputs on plant growth and productivity while minimizing environmental harm. A multi-factorial experiment was conducted using a Controlled Environment Agriculture (CEA) system equipped with IoT-enabled sensors, renewable-energy-powered smart irrigation, and innovative biofertilizers. Real-time monitoring of soil moisture, nutrient status, temperature, and plant health was performed using drones and AI-driven analytics. This integrated system enabled continuous evaluation of crop conditions and timely decision-making. Results showed a significant improvement in the integrated green technology treatment over the control plots. Water usage dropped by 55% and fertilizer consumption by 20%, while overall plant biomass and yield increased noticeably. Additionally, early pest and disease detection through sensor-based and AI-supported monitoring reduced reliance on chemical pesticides. These outcomes confirm that combining precision farming tools with biological inputs and smart monitoring solutions improves crop productivity, resource efficiency, and environmental sustainability. This study presents a scalable, climate-resilient agricultural model. The integration of digital technologies with biological solutions offers a promising pathway for sustainable food production and ecological conservation in the era of climate change.

Increasing frequency and intensity of droughts threaten grassland ecosystems. Semi-natural grasslands vary in age from ancient to younger sites established on former arable land. While species richness and composition are known to affect drought resilience, little is known about how grassland age shapes drought responses through eco-evolutionary processes at the plant population level. We explored how grassland age and local plant-soil adaptation shape the drought resilience of plant populations by reciprocally combining soils and genotypes of a common grass, Briza media, collected from young, intermediate, and ancient grasslands - last cultivated 28-63, 63-84, and > 84-300 years ago, respectively-and subjecting the resulting mesocosms to a drought event. Ancient grassland soils enhanced drought resistance and recovery compared with younger soils. Enhanced drought resilience was primarily explained by lower abundance of putative fungal pathogens in older soils. Plants grown in 'home' soils from their sites of origin were more productive and invested less in root production to withstand water stress, indicating the important role of local plant-soil adaptation. Our results show the long-lasting legacy of land use history in soil microbial communities and their significant role in shaping drought resilience across grassland populations.

Modern agriculture faces increasing pressure from climate change, scarcity of water and arable land, and growing environmental pollution, which limits the effectiveness of traditional soil-based farming. Hydroponic cultivation offers a promising alternative but requires precise, adaptive control of nonlinear processes under varying external disturbances. This paper presents an intelligent control system for hydroponic complexes based on a hybrid Fuzzy–PID strategy integrated with an IoT architecture using the ESP8266 microcontroller. The experimental setup employs a nutrient film technique (NFT) configuration with a distributed sensor network for pH, electrical conductivity (EC), air and solution temperature, humidity, and light intensity. The fuzzy logic module continuously tunes the PID gains in real time according to error and error variation, enhancing robustness and adaptability. Experimental and simulation results demonstrate that the proposed controller maintains pH in the optimal range of 5.8–6.5, improves settling time by approximately 25–30%, and reduces overshoot by about 20% compared with conventional PID control. Moreover, water and nutrient consumption decrease by 10–15% due to more stable regulation of process variables. The IoT-based monitoring subsystem enables reliable real-time supervision and remote management, making the solution suitable for sustainable, resource-efficient agriculture.

Land use can strongly influence aggregate stability, a key factor in conserving soil organic carbon over extended periods. Therefore, the long-term effects of three different land-use types on aggregate stability (AS) and aggregate-associated organic carbon (AAOC) were assessed in this study. The study was conducted on the campus of Çukurova University, Adana, Türkiye. The land-use types were; arable land (AL, continuously used as cropland since 1974), olive orchard (OO, cultivated as olive trees since 1974), and eucalyptus plantation (EP, established in 1997). In June 2025, soil samples were collected from the surface horizon of each land-use type. The results showed that AS and AAOC differed significantly among the long-term different land-use types. In each land-use type, OO and EP significantly affected AAOC. The highest AS was obtained in OO and EP compared to AL, which is conventionally tilled. The land-use type EP had the highest OC (1.48%), which was 32.1% higher than OO (1.12%) and 80.5% higher than AL (0.82%), followed by OO and AL. The highest AAOC was determined in the 4–2 mm aggregate, followed by the 1–2 mm and 0.5–1 mm aggregates, respectively. The results indicated that orchard and plantation promoted improved aggregation and enhanced SOC retention within surface horizon, whereas conventional tillage showed reduced aggregate stability and carbon retention.

Abstract Context The recent decline in biodiversity, which is of particular concern in agriculturally managed landscapes, pushes researchers and practitioners to evaluate potential solutions, such as agricultural diversification. This requires the ability to provide robust conclusions about causal relationships in agroecosystems, which is hindered by limitations of conventional research methods. Findings from controlled plot experiments do not always translate to real-world conditions, while analysis of real-world data is often constrained by methodological difficulties. Objectives The objectives of this study are (1) to apply a novel analytical framework in ecological and agricultural sciences in order to (2) provide robust evidence on ground beetle responses to different aspects of diversification in real-world agricultural landscapes. Methods We applied the propensity-score analytical framework to ecological data collected in the landscape experiment patchCROP on species richness and activity density of carabid beetles. Results The shift from large sole-cropped to small-scale and diversified fields and reduced agricultural management intensity led to a significant increase in species’ richness and activity density of carabids. Edge effects mainly manifested through increased species richness. Locations adjacent to flower strips did not show a significant increase in either response variable on average, but temporal variation of this effect was observed. Conclusion Patch cropping combines the effects of small‑scale crop heterogeneity and high edge density. Combined with less intense agricultural management, it represents a beneficial strategy for supporting carabid beetle species richness and activity density. Further research is needed to explain the effect of flower strips, including sampling within the flower strips and accounting for predation.

In Côte d’Ivoire, soil fertility is a major challenge for the sustainability of agricultural systems. The progressive degradation of arable land is exacerbated by the intensive use of chemical fertilizers, resulting in declining crop yields and increased environmental risks. This study, conducted in Karakoro (Northern of Côte d’Ivoire), aimed to contribute to the development of a soil fertility management strategy for sorghum-based systems by characterizing cropping practices and soil fertilization methods. Data were collected through individual surveys, focus group discussions, and semi-structured interviews. A total of 175 sorghum producers were surveyed regarding cultivated varieties, cropping techniques, constraints, and fertilization practices. The results revealed a strong predominance of red-grained sorghum (99.3%), mainly used for local beer production, household consumption, and marketing. The dominant cropping practices were animal-drawn ploughing and furrow sowing (98%). The main constraints identified included soil infertility (71.7%), climate variability, high input costs, and pest pressure. Fertilization practices relied primarily on mineral fertilizers (NPK and urea), applied in microdoses, while compost use was entirely absent across the study area. However, farmers expressed increasing interest in compost derived from plant residues due to its local availability. These findings highlight the need for simple and context-adapted solutions. The integration of organic fertilizers with agroecological practices could enhance soil restoration and improve the sustainability and resilience of sorghum-based cropping systems in northern Côte d’Ivoire.

Animal-vehicle collisions (AVCs) are an increasing concern globally, yet domesticated animals, particularly dogs, remain understudied. We investigated spatial and temporal patterns of dog-vehicle collisions (DVCs) in Lithuania. Landscape variables such as distance to buildings, forests, meadows, and arable land, as well as land cover composition within a 500 m radius, were analyzed using GIS and compared to randomly generated pseudo-absence points. Temporal patterns were analyzed monthly, daily, and hourly. There was a significant difference in the number of DVCs occurring at sunrise and sunset. Moreover, DVCs were more frequent on weekends, peaking on Fridays and Sundays. Spatially, DVCs were significantly more likely to occur closer to built-up areas and meadows, and farther from forests and arable land, compared to random pseudo-absences, indicating a strong association with human-modified habitats. These findings indicate that DVCs are more influenced by human-modified landscapes and dog activity patterns, particularly around dawn and dusk.

Introduction Relocated farmers in underdeveloped mountain areas face multiple livelihood constraints, making it difficult for them to cope with external risks and challenges. Enhancing the livelihood resilience of these households is essential for achieving stable development and common prosperity in resettlement areas. Methods This study focuses on farmers relocated due to poverty alleviation programs in Guizhou Province. A three-dimensional evaluation framework, comprising buffering capacity, self-organizing capacity, and learning capacity, was developed to assess livelihood resilience. Principal component analysis (PCA) and the TOPSIS model were employed for quantitative measurement. The barrier degree model and geodetector model were used to identify key constraints and core drivers of livelihood resilience. Results The findings indicate that: (1) Over 80% of relocated farming households exhibit medium-to-high livelihood resilience, with only a small proportion showing low resilience. (2) county town resettlement shows significantly higher resilience than town resettlement or central village resettlement. (3) The primary constraints limiting livelihood resilience are housing area, arable land area, and labor force size. (4) A key driver of enhanced livelihood resilience is the synergistic interaction between household income and the social environment of the community. Discussion Relocated farmers have largely achieved livelihood transformation. This study provides a theoretical basis for identifying development bottlenecks and offers insights for optimizing support policies in resettlement areas.

Rapid urbanization, climate instability, shrinking arable land, and increasing food demand are intensifying pressure on global agricultural systems. Conventional farming methods, while foundational to food production, contribute significantly to greenhouse gas emissions, water depletion, soil degradation, and biodiversity loss. In response, smart farming technologies, hydroponics, and vertical agriculture have emerged as controlled-environment production systems that promise land-use efficiency, water conservation, and localized supply chains. However, the environmental sustainability of these systems remains contested due to high electricity demand and embodied carbon emissions. This review critically examines the carbon footprint implications of hydroponic vertical farming compared with conventional open-field and greenhouse systems. It elaborates on greenhouse gas accounting methodologies, life cycle assessment techniques, energy-based emission modeling, and emerging sensor-based carbon monitoring systems. This review suggest that hydroponic vertical farming can significantly contribute to urban food resilience and environmental sustainability when supported by renewable energy transitions and technological efficiency improvements. Rather than replacing conventional agriculture, it functions most effectively as a complementary climate-smart production system.

Archaeoentomology reconstructs past environments and human activities through the analysis of insect remains preserved in archaeological deposits. Although different arthropod groups can be identified when preservation conditions allow, taxonomic resolution often depends on the preservation of diagnostic sclerotized structures, which limits the detection of poorly preserved or very small taxa. Here, we evaluate whether ancient sedimentary DNA (sedaDNA) can complement conventional morphological identifications and provide comparable and reproducible information from the same archaeological samples. We analyzed sediments from two water-saturated Gallo-Roman wells in France using a metabarcoding approach and directly compared the molecular results with morphological identifications made on the corresponding layers. Several groups detected by sedaDNA were absent from the morphological assemblages, despite being ecologically plausible in the studied contexts. Conversely, some beetle taxa, despite being well represented morphologically, were not detected in the DNA data. SedaDNA and morphological data revealed different communities for the two wells. One well exhibited low taxonomic and phylogenetic diversity and was largely dominated by bird mites, generalist decomposers, and other weakly sclerotized arthropods, while the other well had greater overall diversity and was dominated by insects associated with an open environment, likely consisting of arable land. Despite these differences in taxonomic composition, these results highlight the role of local environmental conditions and human practices in shaping distinct arthropod communities at the two archaeological sites. By integrating molecular and morphological data, this comparison demonstrates that sedaDNA does not replace the fossil record but expands the detectable fraction of the arthropod community.

Urban agriculture has become increasingly significant in African cities as a strategy for food security, income generation, and sustainable urban development. This study employed geospatial techniques, anchored in the Sustainable Livelihoods Framework (SLF), to map and analyze market gardening activities and land use changes in Jos South Metropolis, Nigeria, over a ten-year period (2014–2024). The research addresses critical knowledge gaps regarding the spatio-temporal dynamics and sustainability challenges of urban agriculture in Jos Metropolis, where comprehensive data on arable farming activities had been lacking despite extensive crop cultivation. High-resolution satellite imagery with 15 cm spatial resolution was acquired from the National Space Research and Development Agency (NASRDA) for both time periods and processed using ArcGIS 10.8 software. Through systematic on-screen digitization, spatial analysis, and change-detection algorithms, the study quantified changes in agricultural land use, settlements, and water bodies. Extensive fieldwork involved systematic reconnaissance of Jos South Local Government Area (LGA) to identify active market gardening sites, establish ground control points, and conduct interviews with local farmers and agricultural extension agents. Results revealed that market gardens covered 9.59 km² in 2014, declining to 9.03 km² in 2024, representing a 5.84% reduction. Conversely, settlements expanded dramatically from 43.82 km² to 79.32 km², a 34.76% increase, while water bodies decreased marginally from 4.57 km² to 4.46 km², highlighting increasing pressure on water resources driven by urbanization. The study identified key constraints to sustainable urban agriculture including land tenure insecurity, with 63% of dry-season farmers being landless; limited access to irrigation water; pest and disease pressure; inadequate fertilizer supply; and increasing competition for land from urban development. Field surveys documented diverse cropping systems, with farmers cultivating temperate vegetables such as tomatoes, lettuce, cabbage, carrots, and Irish potatoes throughout the year using fadama lands along river channels and mine ponds. The intensive cropping systems documented included plots undergoing three cropping cycles per year, demonstrating both the productivity potential and sustainability challenges. The findings align with global evidence that urban agriculture persists in the face of urbanization, serving multiple livelihood, ecological, and social functions. The research demonstrates the capability of remote sensing and GIS in monitoring urban agricultural dynamics and provides baseline data for urban planning and agricultural policy formulation. The study recommends establishing agricultural zones in fadama areas, strengthening extension services, enacting secure land tenure policies, and fostering rural–urban production synergies to sustain this vital economic sector in the face of rapid urbanization.

While urbanization is accelerating across China’s black soil regions, its specific impact on trace metal accumulation in the vulnerable urban-agricultural transition zones remains poorly understood. This study investigated such a zone in Arongqi, analyzing 32 topsoil samples (0–20 cm) from three subzones for concentrations of Cd, Cu, Pb, Zn, As, Hg, Ni, Cr, Sb, Ba, Co, Mo, Sr, V, Tl, Ag, pH, and soil organic matter (SOM). Ecological and health risks were evaluated using enrichment factors, the pollution load index (PLI), and health risk models. Pollution sources were apportioned via correlation analysis, principal component analysis, and positive matrix factorization. Results showed that urbanization did not exacerbate soil acidification but significantly reduced SOM, resulting in significant enrichment of Cd, Pb, Zn, Ba, Co, Mo, Sr, V, and Ag relative to background levels. The overall pollution status was moderate (PLI), yet distinct spatial patterns emerged: the Urbanizing Zone showed mild contamination by Cd, Ba, and Mo; the Peri-urban Arable Land by Cd, Mo, Sr, V, and Ag; and the Urban Zone by Cd, Pb, Ba, and Sr. The total carcinogenic risk was moderate for both children and adults, with higher risk for adults in the Peri-urban Arable Land and for children in the Urban Zone. Oral ingestion and dermal contact were the dominant exposure pathways, with Cr and As being the key carcinogenic factors for children and Cr also a significant risk for adults. Source apportionment identified five major contributors: traffic emissions (27.46%), soil parent material (24.77%), contemporary agricultural activities (24.54%), urban combustion (11.22%), and historical agricultural practices (11.22%). This study provides a new, spatially-resolved perspective for risk assessment and source management during the urbanization of black soil regions.

Abstract. The formation of organo-mineral associations in the pedosphere are key processes in the global carbon cycle, the mineralization of organic matter, and the release of carbon dioxide into the atmosphere. In this context, the C-sequestration capacity of the reactive fraction of the mineral matrix of post-agrogenic soils was assessed. These soils represent a demutation series of the restorative succession of forest ecosystems: agricultural lands → ruderal stage → meadow stage → shrub stage → sparse woodland of naturally afforested areas of the Ukrainian Carpathians. The calculation of the carbon sequestration potential was performed with consideration of the content of mineral particles sized 0–50 µm, taking into account the characteristics of the particle-size (granulometric) distribution of the studied soils, as well as the conceptual approach applied in this study to divide the organic fraction of soil into a dispersed fraction (>50 µm) and a mineral-associated fraction (<50 µm). It was established that the carbon content in the mineral-associated fraction was 18.8 ± 3.7 g C/kg in the soil under forest, 15.0 ± 0.8 g C/kg at the forest edge, 19.8 ± 0.5 g C/kg in meadow soil, and 14.3 ± 1.2 g C/kg in arable soil. The mineral matrix of soils under meadows and on forest edges is the most C-saturated, while soils under forest and arable land display lower levels of C-saturation. The carbon content in stable aluminum-organic compounds is the highest in forest soil and the lowest in meadow soil. The highest C-sequestration potential of the 0–20 cm layer is characteristic of forest soil – 23.7 t/ha, lower in arable land – 17.0 t/ha, and the lowest in meadow soil – 2.5 t/ha. The C-sequestration capacity of soils in post-agrogenic ecosystems of the Ukrainian Carpathians decreases during the process of silvatization of arable lands due to the accumulation of organic carbon compounds in soils under meadows and forest edges, resulting from the gradual formation of a turf horizon on former arable soils, the decomposition of root residues, and the migration of water-soluble organic matter through the transformation of broad-leaved and coniferous litter. In the soils of post-agrogenic ecosystems, there is an increase in the labile component of organic matter, which at later stages of sylvatization serves as a pool for the formation of stable mineral-associated complexes. That is, at the first stages of natural afforestation (sylvatization), the process of short-term C-sequestration in the form of dispersed (particulate) organic matter prevails, and at the later stages, it occurs as deposition into mineral-associated organic matter. Assessing the potential for carbon binding in the soil provides an opportunity to predict and manage carbon sinks on former arable lands, particularly through the preservation of naturally regenerated forests (spontaneous forests).

Abstract The world’s growing population raises concerns about future food security. At the same time, environmental challenges such as climate change, biodiversity loss, and the depletion of natural resources must be addressed, calling for a transformation of agriculture. This need is exacerbated by the limited availability of land suitable for arable farming. In this study, we examined the potential of agricultural land for arable farming and grassland use in Switzerland, under the premise that agricultural land use should align with the biophysical and environmental capacity of each location. In an iterative co-design process with scientists and public authorities, we elaborated three scenarios for agricultural transformation, which progressively incorporated (i) biophysical constraints (soil, climate, and topography) and (ii) environmental constraints (soil loss and eutrophication due to risk of erosion), as well as (iii) greenhouse gas emissions from drained organic soils. Our results show that the allocation of 40% arable land and 60% grassland in the most restrictive scenario closely resembles the current distribution (46% and 54%), respectively. However, the scenarios also revealed significant spatial shifts between arable land and grassland at the local level: only two-thirds of today’s arable land areas match their natural site conditions. Evidence from this study underscores the critical importance of site-adapted transitions of agricultural land use and the need for site-adapted management alternatives for farmland presently assigned to inadequate land use. Overall, this research provides a novel contribution by allowing the identification of hotspot areas for agricultural transformation at the local scale. We show that these site-specific land use analyses are essential for guiding effective land use planning and policy advice that strengthen the integrity of environmental performance and agricultural productivity, and support the development of targeted and sustainable land use strategies.

From soil carbon towards system sustainability: Integrating SOC modelling and life cycle assessment to evaluate environmental trade-offs in carbon farming

Fungal spores, a common component of biological air pollution, may negatively affect human health by triggering allergic symptoms. To assess the risk of exposure to allergens, abundant and highly allergenic Alternaria spp. and Cladosporium spp. spore concentrations are routinely measured, but almost exclusively on rooftops. In contrast, research on concentrations at ground-level, where people spend most of their time outdoors, is scarce. Although there is evidence that the main source of fungal spores is agricultural areas during harvest, it remains unclear how the proximity of crops influences ground-level spore concentrations. To address this, portable volumetric traps and meteorological stations were used to conduct three ground-level measurement campaigns (before, during, and after harvesting) throughout the main sporulation season in 2023. Nine sampling points in Poznań (Poland) represented the main land use types: agricultural, urban, and forest. The maximum concentrations were 1369 spores/m3 (Alternaria, 47% higher than the rooftop maximum) and 88820 spores/m3 (Cladosporium, 458% higher), both recorded in agricultural areas. On average, concentrations of Cladosporium spores increased during harvest, whereas Alternaria concentrations substantially increased after harvest. The highest concentrations of Alternaria and Cladosporium spores were associated with air influxes from arable land (rho=0.20 and 0.35, respectively) and grasslands (only Cladosporium, rho=0.24). Overall, we successfully established an extensive, mobile, ground-level spore measurement network, which revealed differences in fungal spore concentrations across land-use types, harvest activities, and in comparison with regional-level measurements. This study represents an important step toward assessing airborne fungal spore concentrations relevant to human exposure.

Most agricultural landscapes are composed of a variety of habitats. A landscape perspective is needed to understand biodiversity decline, but many studies focus on single habitat types. In addition, the use of local resources by species within and across habitats implies that species and their habitats are linked in species‐habitat networks. However, studies on these networks are scarce. Here, we used grid‐based sampling to assess wild bees at 224 sampling locations across all major habitat types, that is, arable land, grassland, forest and orchard, in 14 differently composed agricultural landscapes of Southern Germany. We assigned wild bees to habitat types based on the dominant habitat cover surrounding their sampling location to establish species‐habitat networks and assessed how these networks differed in modularity and robustness to habitat loss. Orchards harbored more wild bees than expected based on their proportional cover in the landscape, indicating a preference for this extensively managed but threatened habitat by wild bees. Orchards also supported the highest species richness and proportion of oligolectic wild bees, while forests harbored the lowest richness and more social species. Landscape diversity affected both structure and robustness of bee‐habitat networks in response to the simulated loss of habitats. Networks in more diverse landscapes had higher modularity but tended to be less robust, showing that greater landscape diversity and modularity do not necessarily buffer against the effects of habitat loss. However, this effect appeared to be mainly driven by increases in network size, as standardized modularity and robustness (z‐scores) were not affected by landscape diversity. We could show that species‐habitat networks are a powerful tool to inform ecologists and policy makers about the importance of key habitats and landscape diversity for species conservation. Key habitats for wild bee conservation include extensively managed habitats like traditional orchards. Nevertheless, all habitat types support a similar proportion of endangered species, emphasizing the importance of a diverse landscape. Conserving wild bees requires a variety of complementary habitats at the landscape scale and must consider the management of traditional and intensively managed habitats alike. Policy measures targeting landscape diversity are urgently needed.

Changes in human diets and cropping systems have led to a growing global dependence of agriculture on pollination, but pollinators are experiencing decline worldwide. Key stressors contributing to pollinator decline include natural habitat loss, a reduction in flower resources, and exposure to pesticides. In this study, we investigated effects of land use and estimated agrochemical use (fertilizer, biocides and plant growth regulators) on the abundance and diversity of pollinators across 24 landscape sites in the North China Plain using data from pan trapping across three seasons. We quantified agrochemical use as the average use intensity in agricultural production land (comprising arable land, orchards, agroforestry, and planted trees) and the landscape-wide exposure. The most abundant potential pollinators were bees (mainly Halictidae) and flies (mainly Muscidae and Syrphidae). Pollinator abundance, diversity and species composition were significantly affected by sampling season, with higher abundance and species diversity in autumn than in spring and summer. Pollinator abundance was positively associated with the proportion of arable land but negatively related with the proportion of orchard land and the agrochemical use intensity in agricultural production land and to the landscape-wide agrochemicals use. Overall, this study uncovered a weak signal of agrochemical use but a strong effect of season on pollinator diversity in the landscape setting of the North China Plain. Lower agrochemical use and a greater proportion of arable land were associated with higher abundance of some pollinator species in this intensive cereal production region. Results suggest that in the context of the North China Plain, pollinator abundance might be shaped more by aphid populations that support dominant species of hoverflies and by bare soil that favours nesting of wild bees, rather than by floral resources. Findings suggest that reducing agrochemical use intensity in agricultural production land, especially in orchards and other systems with trees, might facilitate pollinator conservation.